Changepoint label geom

Draws labelled regions behind a series — the central object of supervised changepoint detection (Hocking et al.), where an expert marks intervals as containing a change or not and the penalty is learned from those labels. Fill defaults to the label's change status, so a plot of labels reads as a picture of what the expert asserted; map fill to the status column of cpt_label_error() instead to get the correct / false-positive / false-negative display.

Usage

geom_cpt_label(
  mapping = NULL,
  data = NULL,
  ...,
  alpha = 0.25,
  colour = NA,
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = FALSE
)

Arguments

mapping

Aesthetic mappings. Requires xmin and xmax; ymin/ymax default to the panel extent. fill is commonly mapped to change or status.

data

A data frame of labels, e.g. cpt_labels() output.

...

Other arguments passed to ggplot2::geom_rect().

alpha

Fill transparency. Defaults to 0.25.

colour

Border colour. Defaults to NA (no border).

na.rm

If FALSE, missing values are removed with a warning.

show.legend

Whether to show a legend.

inherit.aes

Whether to inherit the plot's aesthetics. Defaults to FALSE.

Value

A ggplot layer.

See also

cpt_labels(), cpt_label_error(), scale_fill_cpt_label().

Other ggplot2 layers: geom_changepoint(), geom_cpt_ci(), geom_cpt_event(), geom_cpt_region(), geom_cpt_segment(), stat_changepoint()

Examples

library(ggplot2)
set.seed(2026)
d <- data.frame(t = 1:100, y = c(rnorm(50), rnorm(50, 4)))
labs <- cpt_labels(c(40, 70), c(60, 90), c("change", "no_change"))
ggplot(d, aes(t, y)) +
  geom_cpt_label(aes(xmin = start, xmax = end, fill = change),
                 data = labs) +
  geom_line() +
  scale_fill_cpt_label()